[ 130 ]

NEUROLOGICAL STUDIES OF THE CARDIAC VENTRICLES OF MAMMALS

By FRANCIS DAVIES, E. T. B. FRANCIS AND T. S. KING Departments of Anatomy and Zoology, University of Sheffield
INTRODUCTION

The present histological study of the cardiac ventricles of mammals has three main objects: first, to determine whether or not nerve cells are present; secondly, to observe the types of nerve endings related to the various tissues comprising the ventricles; and thirdly, to investigate the nerve relationships of the atrioventricular part of the cardiac conducting system (the A.v. node and the A.v. bundle and its branches). Subsequent to the description by Scarpa (1794) and Lee (1849a, b) of naked-eye 'ganglia' on the epicardial nerves in the ventricles of the cow, horse and man, and the demonstration microscopically by Remak (1838, 1839, 1844) of nerve cells on some of the ventricular nerves in the calf, sheep, pig and man, numerous workers have studied this question. The relevant literature, extensively reviewed by Michailow (1912), Perman (1924), and Stohr, jun. (1928), reveals that there are wide discrepancies in the observations, even in the results of the studies of the same species of mammal by different investigators. Most observers maintain that cardiac nerve cells are limited to the atria. Only a few describe a widespread distribution of ventricular nerve cells, situated mainly in the epicardium, e.g. Kasem-Beck (1885, 1887) in the dog, calf, sheep and pig; Smirnow (1895, 1905) in the rabbit, hare, squirrel, wolf, dog, cat, calf and man; Valedinsky (1905, 1908, 1910) in the rabbit, dog, calf, pig and man; and Michailow (1912) in the mouse, rat, rabbit, dog, sheep, pig and ape. Smirnow, Valedinsky and Michailow are agreed that the nerve cells are principally multipolar, few being bipolar or unipolar, while Kasem-Beck found many bipolar and unipolar cells as well as multipolars. Some observers have described organized sensory nerve endings in the ventricular epicardium and endocardium in the form of complicated branchings (Smirnow, 1895), encapsulated arborizations and coils (Michailow, 1912), large diffuse endings (Nettleship, 1936), or tactile corpuscles (Landau, 1950), while others (Fukutake, 1924; Edens, 1924; Woollard, 1926) observed only endings of simple type. Huber & DeWitt (1898) and Fukutake (1924) failed to find any motor end-plates or muscle spindles in relation to the ventricular muscle fibres like those in skeletal muscle, whereas King (1939) described both of these types of endings, and Lavrentiev (1946) reported that Plechkova, working in his laboratory, found proprioceptor endings similar to those in skeletal muscle, as well as true spirals around the myocardial fibres. Woollard (1926), Jones (1927), Boeke (1932) and Tcheng (1951) maintained that nerves penetrate into the substance of the muscle fibres and end near their nuclei. Several workers have noted that the A.v. bundle is accompanied by nerves, some stressing that these nerves vary considerably in number in different mammals.

Neurological studies of the cardiac ventricles of mammals

131

Some writers have suggested that it may be the nerves, and not the bundle itself, which conduct the impulse for cardiac contraction from the atria to the ventricles (Agduhr, 1933; Wahlin, 1935; Glomset & Birge, 1945, 1948; Field, 1951).
MATERIALS AND METHODS

The hearts studied comprised representatives of many orders of mammals, as follows, the number of hearts examined from each species being indicated in brackets. ORNITHODELPHIA: platypus (Ornithorhynchus) (1), echidna (Pro-echidna) (2). DIDELPHIA: wallaby (Macropus robustus) (2), kangaroo (M. giganteus) (1). MONODELPHIA: Insectivora: hedgehog (Erinaceus) (1), mole (Talpa) (2); Rodentia: rat (Mus) (3), rabbit (Oryctolagus) (4), guinea-pig (Procavia) (4); Carnivora: cat (Felis) (3), dog (Canis) (4); Cetacea: porpoise (Phocoena) (2); Artiodactyla: Ruminants-cow (Bos) (6), sheep (Ovis) (4), red deer (Cervus) (2); Non-ruminantspig (Sus) (3); Perissodactyla: horse (Equus) (2); Chiroptera: common bat (Vespertilio) (2), fruit bat (Pteropus) (1); Menotyphla: tree shrew (Tupaia) (1); Primates: (Homo) (adult, 3; child, 4). In most cases the hearts were serially sectioned in the transverse plane. In the case of the horse, cow, red deer and pig, and in two of the adult human hearts, only blocks of tissue were sectioned. The following were the principal techniques used: impregnation in bulk by a silver-pyridine method (Blair & Davies, 1935), or staining of individual sections by haematoxylin and eosin, or iron haematoxylin and picrofuchsin, or by the silver technique of W. Holmes (1947). In addition, in many cases frozen sections of blocks of tissues were treated by the Bielschowsky-Gros technique.
OBSERVATIONS

Nerve cells. Only in Artiodactyla and Cetacea were ventricular nerve cells found to be abundant, lying for the most part in the epicardium, and only in these animals were numerous nerve cells observed in relation to the A.v. bundle and its limbs. No nerve cells were seen in the endocardium or myocardium in any heart examined, with the exception of the cow in which a few were found in the superficial part of the myocardium. The widespread distribution of epicardial nerve cells in the calf is shown in PI. 1, figs. 1 and 2, each dot representing a group of cells ranging from about six up to fifty or more. Their positions were determined by graphic reconstruction from serial sections and plotted on photographs of another heart. The cells extend as far as the apex of each ventricle, on both ventral and dorsal surfaces. Many of the cells lie in the course of nerve trunks that are related to the coronary arteries, but some are situated quite apart from the vessels. In the cow and calf epicardial nerves are visible to the naked eye through the intact epicardium and discrete swellings on them, similar to the 'ganglia' described by Scarpa and Lee, could be seen in places where the nerves crossed over or under coronary arteries, though some have no relation to the vessels. Histological examination of several of the swellings showed that, while many are true ganglia, others contain no nerve cells and are produced by a localized thickening of the epineurium. The disposition of endocardial nerve cells in the calf is shown in PI. 1, figs. 3 and 4; the dots have the same significance
Anatomy 86
9

132

Francis Davies, E. T. B. Francis and T. S. King

and were plotted in the same way as those showing the epicardial cells. They were found to be limited to the A.V. bundle and its limbs (P1. 2, fig. 9). In the adult cow and in the sheep, nerve cells were found to be numerous and to be distributed similarly to those in the calf. In the red deer, numerous epicardial nerve cells extend over the proximal three-quarters of both ventricles, but none were found in relation to the A.v. bundle or its limbs. In the pig, epicardial nerve cells extend only for a short distance distal to the coronary sulcus, though in this restricted ventricular territory they are fairly numerous, but none were seen in relation to the A.v. bundle or its limbs. So far as the present work goes, therefore, there appears to be a contrast between the extent of distribution of ventricular nerve cells in Ruminants (cow, calf, sheep and red deer) and Non-ruminants (pig) amongst the Artiodactyla. In the horse, as a representative of the Perissodactyla, by marked contrast with the Artiodactyla, no nerve cells at all were found in the ventricles. In the porpoise, nerve cells were found in great numbers, with a distribution similar to those in the cow, calf and sheep. In the other animals no nerve cells were found in the ventricles, except in the mole, common bat and fruit bat in which a few epicardial cells were seen to extend for a very short distance beyond the coronary sulcus, and in one of the four dogs examined, in which a small group of epicardial cells was seen on the posterior surface of the left ventricle near the coronary sulcus. In Artiodactyla and Cetacea the nerve cells are frequently arranged in the form of ganglia alongside nerve trunks with nerve fibres passing between the ganglia and the trunks (PI. 2, fig. 6); in places they form ganglia within the substance of nerve trunks (PI. 2, fig. 5), and in others a few cells or even single cells occur in the course of nerves. The epicardial nerve cells are mostly bipolar or unipolar (P1. 2, fig. 10) in type, few being multipolar, whereas those related to the A.v. bundle and its limbs are principally multipolar, few being bipolar or unipolar. In general, the multipolar cells have numerous processes (P1. 2, fig.11), most of which enter neighbouring nerve trunks but could not be traced in the sections to their individual destinations; some have, in addition, short processes which end near the cell body in club-like enlargements. In most cases the axon could not be identified, though sometimes a process much thinner than the others was observed which probably represented the axon. The processes of the bipolar and unipolar cells are usually thick. The two processes of the bipolar cells frequently arise from opposite poles of the cell body (PI. 2, fig. 8); sometimes they arise conjointly from one pole (PI. 3, fig. 15). Occasionally the single process of a unipolar cell was observed to divide into two some distance from the cell body (P1. 2, fig. 12), such cells thus resembling pseudo-unipolar cells of spinal and cranial nerve ganglia. This appearance suggested the possibility that the unipolar and bipolar cells might be sensory (afferent) in nature. Frequently the endings of nerve fibres, in the form of small knobs, were seen in relation to the cell bodies and processes of the multipolar cells (PI. 3, figs. 13, 14), whereas no nerve endings were found in relation to the bipolar and unipolar cells. These observations suggested that the bipolar and unipolar cells might be afferent and the multipolar cells efferent in nature. In general, individual nerve cells are surrounded by a capsule of satellite (neurilemmal) cells (P1. 2, fig. 12), and the nerve endings related to the multipolar cells lie both outside and inside the capsules. Nerve endings. The observations on the types of nerve endings in the ventricles

Neurological studies of the cardiac ventricles of mammals

133

were made chiefly on the hearts of the cow, calf and dog. Preparations made by the prolonged osmic acid method of Kiss (1932) showed numerous medullated nerves in the epicardium (P1. 2, fig. 7) and to a lesser extent in the endocardium and myocardium. Their myelin sheaths vary considerably in thickness and the fibres themselves also vary in diameter. Numerous non-medullated nerves are present in the epicardium and endocardium and many pass from these sites into the myocardium where they form the myocardial plexus. As they approach their terminations all the nerve fibres examined were non-medullated, and it appeared that the medullated nerves lose their myelin sheaths some distance from their terminations. The nonmedullated fibres present varicosities along their course, some in the form of spherical knobs, others as elongated fusiform or irregularly cylindrical swellings. In histological sections, these structures, where the nerves surround groups of nerve cells, present a false appearance of knob-like nerve endings, and it is suggested that this may account for the description by other authors of nerve endings in relation to all types of nerve cells. Many neurilemmal cells (Schwann cells) are related to the non-medullated fibres. The only type of nerve ending found in the epicardium and endocardium was a simple bulb-like enlargement or a gradual thinning leading to an abrupt termination (PI. 3, fig. 16); no organized endings were seen. No nerve endings resembling the motor endings or muscle spindles of skeletal muscle were observed in relation to the myocardial fibres. It was noted, however, that fine non-medullated nerves ended by forming spirals around individual muscle fibres (PI. 8, fig. 17). No nerve endings were found in the connective tissue between groups of myocardial fibres. The coronary arteries in the epicardium and myocardium were observed to be accompanied by numerous nerve fibres which course both in the adventitia of the vessels and in the surrounding connective tissue. Simple endings of non-medullated nerves, in the form of very small end-bulbs or a tapering off and sudden ending, were noted in the adventitia of the arteries (P1. 4, fig. 25), both in the epicardium and myocardium. The capillary vessels were seen to be accompanied by delicate non-medullated nerves and the impression was gained that they comprise the main components of the plexus of nerves in the myocardium, though it was not possible to be certain since the capillary walls were only infrequently and capriciously impregnated and rendered visible by the silver techniques employed. It was clear, however, that the nerves which left the A.v. bundle (see later) to pass into the ventricular septum are far too few to contribute materially to the very numerous fibres that enter the septum and the ventricular walls from the epicardium along with branches of the coronary arteries. It was a striking feature that, under the microscope, considerable areas of the ventricular muscle were devoid of nerves, and it is believed that this appearance is not due to any vagaries of the techniques. Fine non-medullated nerves were also seen to pass into the media of the arteries to end as slight enlargements or simply by coming to a sudden end, either between the plain muscle fibres or within them (PI. 4, figs. 24, 26). It was noted that only in occasional parts of an artery could such nerves be seen passing into the media. Nerves of the atrioventricular bundle. In Artiodactyla and Cetacea the A.v. bundle was found to be accompanied by numerous nerves (P1. 4, fig. 23), many of which continued along its two limbs. In the other mammals, although many nerves were
9-2

134

Francis Davies, E. T. B. Francis and T. S. King

seen in relation to the A.v. node, relatively few accompanied the A.v. bundle (P1. 4, figs. 20, 21), and the limbs of the bundle contained extremely few, the distal part of the limbs containing practically none. In the dog, for instance, no nerves at all were found in the limbs of the bundle in specimens in which the nerves in the A.v. node and bundle and other parts of the heart were very well impregnated with silver (PI. 4, fig. 22). In Artiodactyla and Cetacea many nerves leave the bundle and its limbs to pass into the ventricular septum, and even in the other animals a number of the nerves related to the bundle were seen to do likewise. In all cases the nerves in the bundle are predominantly non-medullated, few being medullated with thin myelin sheaths. They lie in the connective tissue round the bundle as well as between groups of the muscle fibres comprising the bundle. Nerve endings in the adventitia and media of coronary arteries accompanying the bundle were observed, similar to those described above in the case of the arteries in the myocardium. Two types of nerve endings were noted in relation to the muscle fibres of the bundle. First, a fine plexus was seen on the surface of the muscle fibres (PI. 3, fig. 18); it is best developed about the large Purkinje fibres in Artiodactyla and Cetacea, while in the other animals fine nerves lie on the surface of the bundle muscle fibres though a plexus formation is not well developed. No branches from the plexus were observed to penetrate into the substance of the muscle fibres, and it is suggested that these perifibrillar nerves may be sensory in nature. In places the nerves surround individual muscle fibres but often they are related only to groups of fibres, and not all the fibres or groups of fibres are surrounded by such nerves. Secondly, simple nerve endings in the form of terminal knobs were seen on the surface of some of the bundle muscle fibres (PI. 3, fig. 19), which were epilemmal in position and which it is suggested may be efferent (motor) in nature. Fine non-medullated nerves were also observed in the endocardium close to the peripheral subendocardial Purkinje fibres, but they were not seen to end in particular relationship to the Purkinje fibres themselves. These nerves have exactly the same features as subendocardial nerves that are not related to Purkinje fibres, they end in the same manner and it is believed that they belong to the general system of subendocardial sensory nerves. Likewise, no nerve endings were seen in relation to the Purkinje fibres that, in Artiodactyla and Cetacea, penetrate into the myocardium from the endocardium. In all the animals numerous nerves, mostly non-medullated, are related to the A.v. node, and two types of nerve endings were seen in relation to the nodal fibres, similar to those described above related to the muscle fibres of the A.v. bundle. In all cases numerous nerve cells lie in the atrial septum, close to, but not actually in, the substance of the node; they are multipolar (principally), bipolar and unipolar in type, and processes from the cells pass into the node.
DISCUSSION

Among representatives of many orders of mammals, only in Artiodactyla and Cetacea were ventricular epicardial nerve cells found to be numerous and extensive in distribution, a finding contrary to the results of some previous workers who found few or none in the porpoise (Eiger, 1909a, b) calf (Skworzow, 1874; Perman, 1924), sheep, goat and pig (Perman, 1924). Undoubtedly, there is a variation in the degree

Neurological studies of the cardiac ventricles of mammals

135

to which nerve cells in the other mammals may extend beyond the coronary sulcus, as witnessed by the finding of a few ventricular epicardial cells in one of the dogs examined, and it is believed that such variation accounts for other workers finding some ventricular cells in mammals in which none were found in the present work (Schklarewski, 1872; J. Dogiel, 1877, 1895, 1914; Vigrlal, 1881; Aronson, 1886; Weinreich, 1888; Jacques, 1894; A. S. Dogiel, 1899; Noc, 1899; Fahr, 1909, 1910; Muller, 1911). It is submitted that this widespread distribution of ventricular nerve cells in Artiodactyla and Cetacea is a true morphological difference between these two orders and the others, since no functional adaptation is apparent. The absence of ventricular nerve cells in the horse as a representative of the Perissodactyla, contrary to the condition in Artiodactyla, lends support to the modern tendency, based on other morphological differences, to discontinue the term 'Ungulates' as including both these groups and to raise each to equivalent ordinal rank. Simpson (1945) noted the difficulty of assigning a place to the Cetacea in a scala naturae of mammals. Boyden & Gemeroy (1950) by precipitin tests showed that the serum proteins of Cetacea resemble those of Artiodactyla more than those of any other existing order of mammals (all except the Lagomorpha being included in their tests). The present finding regarding ventricular nerve cells is a further indication of the close relationship between these two orders. Amongst the Artiodactyla, the Ruminants show a much more extensive distribution of ventricular nerve cells than the Non-ruminants, and in this respect they resemble the Cetacea, a feature that supports other morphological evidence of affinity between these two groups. Numerous nerve cells were observed in relation to the A.v. bundle only in the porpoise, calf, cow and sheep, and this agrees with the findings of Tawara (1906) in the calf, Wilson (1909), Meiklejohn (1913, 1914) and A. H. Holmes (1921) in the calf and sheep, Scaglia (1927) and Blair & Davies (1935) in the cow, Glomset & Glomset (1940) in the sheep, and Engel (1910) and Perman (1924) in 'Artiodactyla'. Their presence was denied by Morison (1912) in the sheep, and by Truex & Copenhaver (1947) in the calf and sheep. Wilson (1909) and Glomset & Glomset (1940) described them in the pig, Lawrentjew & Gurwitsch-Lasowskaja (1930) in the rat, and Stotler & McMahon (1947) in the dog. Akkeringa (1949) described numerous 'interstitial cells', which he claimed are nerve cells, in relation to the peripheral subendocardial Purkinje fibres in the cow and horse, but no undoubted nerve cells, comparable in size and structure with those described above, were found in these sites by the present workers. Woollard (1926) and Lawrentjew (1929), on the basis of degeneration studies following excision of sympathetic or vagal nerves, and Nonidez (1939), who noted that sympathetic cells stain less deeply than parasympathetic with his modification of Cajal's chloral-hydrate-silver technique, concluded that all cardiac nerve cells are efferent parasympathetic in nature. The present writers, on purely morphological grounds, suggest that the multipolar ventricular nerve cells may be efferent, and the bipolar and unipolar cells afferent in function. Similar types of nerve cells were noted in the atria, and the same suggestion is made for them. It has not been possible to make out any connexions between the bipolar or unipolar cells and the multipolars, nor to trace the processes of individual cells to their destination, so that the nerve endings described could not be related to the particular types of nerve cells.

136

Francis Davies, E. T. B. Francis and T. S. King

The simple nerve endings observed in the epicardium and endocardium are most probably sensory in function. Lawrentjew (1927) and Nettleship (1936) claimed that the nerve endings in the endocardium degenerate after section of the vagus, and Lavrentiev (1946) by section of various nerves (vagal, sympathetic, spinal) concluded that the sensory nerves in the endocardium are vagal and those in the epicardium sympathetic. Weddell & Zander (1950) found that simple nerve endings in the cornea could be damaged during histological preparation so that they become distorted and made to resemble complicated endings. Such as explanation may account for some workers (Smirnow, 1895; Michailow, 1912; Nettleship, 1936; Landau, 1950) describing complicated endings in the epicardium and endocardium. Tcheng (1951) described what he considered to be sensory nerve endings, in the form of simple knobs, between the atrial muscle fibres in the dog, but in the present work, no nerve endings were found in the connective tissue between the ventricular myocardial fibres. It is, therefore, suggested that the simple endings observed in the adventitia of the coronary arteries, which are undoubtedly sensory in nature, may be the ones concerned in the production of cardiac pain. If the mechanism of production of cardiac pain involves ischaemia of the myocardium with the accumulation of pain-producing metabolites that stimulate pain nerve endings (Lewis, 1947), it would appear that such metabolites might diffuse into the adventitia of the arteries and affect these nerve endings. In this connexion, it is noteworthy that, in dogs, traction on a ligature around a coronary artery produces evidence of referred pain (Sutton & Lueth, 1930), whereas puncturing the heart wall appears to be painless (Singer, 1926; Sutton & Lueth, 1930). Stdhr jun. (1932) could not find nerves in the media of arteries. Woollard (1926) and Blair & Davies (1935) found nerve endings in the media of the coronary arteries, while Boeke (1932) noted that in arteries a single nerve ending could only now and then be found between the muscle cells, and remarked that such an appearance is due to vagaries of the techniques employed. The present writers also noted that the nerve endings in the media of the coronary arteries were only to be found in occasional parts of an artery, and submit that this may be the actual method of innervation of the vessels. Nonidez (1939, 1943) claimed that his silver technique enabled sympathetic nerves to be distinguished from parasympathetic, and maintained that the atria are supplied by the parasympathetic and the ventricles by the sympathetic. No such distinction could be made in the present study. Plechkova (1936) and Lavrentiev (1946) maintained that in their terminal parts one and the same neurilemmal sheath contains a sympathetic and a parasympathetic axis cylinder, and that the antagonistic action of these two parts of the autonomic nervous system on the heart muscle is thus brought about peripherally, i.e. on the same muscle fibre. Such an arrangement was not observed in the present work. It is suggested that the spiral nerve endings around the myocardial fibres described above may be of the nature of sensory stretch receptors. If this be so, the fact that they are not extremely numerous may be attributed to the syncytial nature of the myocardium. Stretch receptors in the ventricular myocardium have been postulated by previous workers on the basis of a reflex fall of blood pressure and slowing of the heart rate following upon a rise of pressure in the left ventricle

Neurological 8tudie8 of the cardiac ventricle of mammals

137

(Jarisch & Richter, 1939; Richter & Amann, 1940), raising the pressure in the ascending aorta (Daly & Verney, 1926), or injection of veratrum alkaloids into various cardiac chambers or into branches of the coronary arteries (Dawes, 1947). Many workers have noted that the A.v. bundle is accompanied by nerves, which are mainly non-medullated; e.g. Tawara (1906), Retzer (1908), DeWitt (1909), Wilson (1909), Engel (1910), Morison (1912), Meiklejohn (1913,1914), A. H. Holmes (1921), Fukutake (1924), Woollard (1926), Scaglia (1927), Wolhynsky (1928), Lawrentjew & Gurwitsch-Lasowskaja (1930), Blair & Davies (1935), Vitali (1937), Glomset & Glomset (1940), Nonidez (1943), Glomset & Birge (1945, 1948), Truex & Copenhaver (1947), Stotler & McMahon (1947), Akkeringa (1949) and Field (1951). Some of the above have noted that the number of nerves varies considerably in different mammals, but all who have studied the Artiodactyla agree that in them these nerves are numerous. Only few nerves were found in the bundle in man by DeWitt, Meiklejohn, A. H. Holmes and Blair & Davies, and in the dog and cat by DeWitt and Meiklejohn, while no nerves at all were seen in man by Fahr (1910) or in the dog by Tcheng (1951). Field (1951), on the other hand, remarked that the richness of the nervous component of the bundle in the dog is striking and is of special interest as this animal has been so widely used experimentally in work on the conduction of the cardiac impulse. The plexus of nerves on the surface of the bundle muscle fibres observed in the present work has also been noted by Retzer, Wilson, Meiklejohn, Woollard, Blair & Davies, Vitali and Akkeringa. It is suggested that these nerves are sensory in function, a belief expressed also by Wilson, Vitali and Blair & Davies, while Meiklejohn and Akkeringa maintained that branches from the plexus penetrate the muscle fibres, the latter author referring to the places of penetration as the 'innervation points' of the muscle fibres. It is suggested that the simple knob-like nerve endings on the surface of the bundle muscle fibres are efferent (motor) in function. Truex & Copenhaver found only simple nerve endings in relation to the bundle muscle fibres and maintained that they lie both on the surface of, and inside the muscle fibres, while Blair & Davies maintained that these endings were hypolemmal in position. Scaglia found nodose endings of nerves on the surface of, and inside, the bundle muscle fibres, as well as endings in the form of various configurations, some encapsulated, in the connective tissue between the muscle fibres. Nonidez described nerve endings in the form of rings, club-shaped enlargements and reticulated swellings on the surface of, and between, the muscle fibres, and maintained that the bundle and the proximal parts of its limbs are supplied by the parasympathetic, though he could not decide whether they are also supplied by the sympathetic since these nerves are not stained by his technique. Woollard concluded that it is highly probable that the bundle is supplied by both sympathetic and parasympathetic nerves. Several workers, reviewed by Davies & Francis (1946), on the basis of cutting, ligurating or compressing the A.v. bundle and thereby producing dissociation between atrial and ventricular contraction, concluded that the bundle conducts the impulse for cardiac contraction, though some writers have stressed that in experimental lesions involving the bundle (Cohn & Trendelenburg, 1910; Glomset & Birge, 1948) the nerves accompanying the bundle are of necessity involved. Obviously the experiments are not 'crucial' in determining which of the two components of

138

Francis Davies, E. T. B. Francis and T. S. King

the bundle, muscle or nerves, conducts the impulse. Agduhr & Strenstr6m (1928) and Agduhr (1933) noted that, in mice, cod-liver oil produced focal areas of myocardial degeneration that may involve the A.v. bundle without producing heartblock, and, as apparently undamaged nerves in the bundle were seen in one case, Agduhr suggested that it may be the nerves that are the true conducting elements. Wahlin (1935) also observed sclerotic patches involving the bundle without producing heart-block but only a prolongation of the normal P-R interval in the electrocardiogram. Field (1951) was impressed by the presence of nerves in the bundle in several mammals examined by him and quotes the work of Agduhr and Wahlin as casting serious doubts on the myogenic theory of cardiac conduction. Glomset and his colleagues (Glomset & Glomset, 1940; Glomset & Birge, 1945, 1948) could find no anatomical evidence, in the form of an A.v. node and bundle, in the dog, rhesus monkey and man, to support the myogenic theory of cardiac conduction, postulating that the cardiac musculature is under the control of the intrinsic nerve cells and nerves. In the case of the dog (PI. 4, fig. 22) and man, the present writers disagree with these observations, and Baird & Robb (1950) have described the A.V. node and the bundle and its limbs in the dog as exactly comparable with these structures in Ungulates. In the present work, the nerves accompanying the A.v. bundle and its limbs were found to be relatively few in number, except in Artiodactyla and Cetacea. Many of them were noted to leave the bundle in the early part of its course to pass into the myocardium of the ventricular septum, an observation also made by Scaglia (1927) and Field (1951), but it appears to the present writers that these nerves are far too few to contribute materially to the plexus of nerves in the septum, the latter being mainly derived from the epicardial nerves. This observation, coupled with the failure to find any motor nerve endings in the myocardium, leads to doubt as to whether the nerves accompanying the bundle can be concerned with the conduction of the impulse from atria to ventricles. Further, if these nerves did convey the impulse to the septal musculature, it might be expected that the upper part of the septum would be activated first, since the earliest nerves to leave the bundle enter this part of the septum, and that the process of activation would extend downwards through the septum. The extensive investigations in dogs by Sodi-Pallares, Rodrigues, Chait & Zuckermann (1951), however, show that activation of the ventricular septum isfrom below upwards, and that the times of arrival of the waves of activation at different points on the right and left surfaces of the septum are in conformity with the distribution of the branches of the two limbs of the bundle. In the present work it was found that the-bundle in the dog contains relatively few nerves and that many of these leave the bundle to pass into the upper part of the septum, while the limbs of the bundle contain no nerves at all. It is submitted that these observations, considered in conjunction with those of Sodi-Pallares et al., are strongly in favour of the conduction of the impulse from atria to ventricles by the bundle and its limbs. Like Nonidez (1943), Truex & Copenhaver (1947), Stotler & McMahon (1947) and Tcheng (1951), the present writers found no nerve endings in relation to the peripheral Purkinje fibres, and, as no nerve endings were seen in relation to the ventricular myocardium that could be interpreted as motor endings, they agree with the

Neurological studies of the cardiac ventricles of mammals

139

opinion of Stotler & McMahon that the relation of the muscle elements of the A.V. bundle and its branches to the ventricular myocardium is best explained by the orthodox conception of the myogenic propagation of the cardiac impulse from atria to ventricles. Collateral evidence, derived from a comparative study of electrocardiograms, that the rate of propagation of the cardiac impulse throughout the ventricles in a number of mammals (and birds) can be correlated with the degree of histological differentiation of the A.v. bundle and its branches, will be presented in another paper by two of the present authors (Davies & Francis).
SUMMARY

1. Only in Cetacea and Artiodactyla, among representatives of a large number of Orders of mammals, were nerve cells found to be numerous and extensive in distribution in the ventricles. This resemblance between these two Orders is adduced as further support for their close systematic relationship. 2. Unipolar, bipolar and multipolar cells were found in all the hearts. From the morphological appearances and the presence or absence of nerve endings in relation to the cells, it is suggested that the multipolar cells may be efferent and the unipolar and bipolar cells afferent in function. 3. Only simple nerve endings were found related to the various tissues comprising the ventricles. It is suggested that the spiral nerve endings around ventricular myocardial fibres are sensory stretch receptors and that the simple nerve endings in the adventitia of the coronary arteries may be concerned with the cardiac pain mechanism. 4. Nerves related to the atrioventricular bundle are described and reasons are given for believing that they are not concerned with conducting the impulse for cardiac contraction from atria to ventricles.

We are indebted to our technical assistants, J. H. Kugler and J. Morrill, for the preparation of the specimens and the photomicrographs. The expenses involved in the work were defrayed by a grant from the Medical Research Council to one of us (F. D.) and grants from the University of Sheffield Medical Research Fund, and for these we wish to express our thanks.
REFERENCES AGDUHR, E. (1933). Die Bedeutung der Diat fur Regeneration von Herzmuskelschaden. Uppsala LdkForen Forh. 39, 65-89. AGDUHR, E. & STRENSTROM, N. (1928). The appearance of the electrocardiogram in heart lesions produced by cod liver oil treatment. Acta paediatr., Stockh., 8, 493-610. AKKERINGA, L. J. (1949). The nervous system of the Purkinje fibres in the heart. Acta Neerl. Morph. 6, 289-299. ARONSON, H. (1886). Beitrage zur Kenntnis der centralen und peripheren Nervenendigungen. Inaug. Diss., Berlin. (Cited by Michailow, 1912.) BAIRD, J. A. & ROBB, J. S. (1950). Study, reconstruction and gross dissection of the atrioventricular conducting system of the dog heart. Anat. Rec. 108, 747-763. BLAIR, D. M. & DAVIES, F. (1935). Observations on the conducting system of the heart. J. Anat., Lond., 69, 303-325. BOEKE, J. (1932). Nerve endings motor and sensory. In Penfield's Cytology and Cellular Pathology of the Nervous System, vol. I, sect. 6, 241-315. New York: P. Hoeber. BOYDEN, A. &. GEMEROY, D. (1950). The relative position of the Cetacea among the orders of Mammalia as indicated by precipitin tests. Zoologica, N.Y., 35, 145-151.

140

Francis Davies, E. T. B. Francis and T. S. King

COHN, A. E. & TRENDELENBURG, WV. (1910). Untersuchungen zur Physiologie des tbergangbundels am Saugetierherzen, nebst mikroskopischen Nachprufungen. Pfiug. Arch. ges. Physiol. 131,
1-86. DALY, I. DE B. & VERNEY, E. B. (1926). The localisation of receptors involved in the reflex regulation of the heart beat. J. Physiol. 62, 330-340. DAVIES, F. & FRANCIS, E. T. B. (1946). The conducting system of the vertebrate heart. Biol. Rev. 21, 173-188. DAWES, G. S. (1947). Studies on veratrum alkaloids. VII. Receptor areas in the coronary arteries and elsewhere as revealed by the use of veratridine. J. Pharmacol. 89, 325-342. I)EWVITT, L. M. (1909). Observations on the sino-ventricular connecting system of the mammalian heart. Anat. Rec. 3, 475-497. I)OGIEL, A. S. (1899). Zur Frage uber den feineren Bau der Herzganglien des Menschen und der Saugetiere. Arch. mikr. Anat. 53, 237-281. L)OGIEL, J. (1877). Die Ganglienzellen des Herzens bei verschiedenen Thieren und beim AIenschen. Arch. mikr. Anat. 14, 470-480. IDOGIEL, J. (1895). Die vergleichende Anatomie, Physiologie und Pharmakologie des Herzens. Kazan. (Cited by Mlichailow, 1912.) DOGIEL, J. (1914). Die Anordnung und Funktion der Nervenzellen des Herzens des Alenschen und der Tiere und ihre Verbindungen mit dem sympathischen, den cerebralen und spinalen Nerven. Pflug. Arch. ges. Physiol. 155, 351-390. EDENS, E. (1924). Wesentlich erweiterte Auflage des vegetativen Nervensystems. In L. R. AIiiller's Die Lebensnerven. Berlin: Springer. (Cited by Fukutake, 1924.) EIGER, M. J. (1909a). Topographic des ganglions du coeur chez le cobaye, chez la souris blanche et chez l'homme. C.R. Soc. Sci. Varsovie. (Cited by AMichailow, 1912.) EIGER, M. J. (1909b). Die Topographie der intracardialen Nervenganglien des Meerschweinchens, der weissen Maus und bei Menschen. Warsaw. (In Russian. Cited by Valedinsky, 1910.) ENGEL, J. (1910). Beitrage zur normalen und pathologischen Histologie des Atrioventrikularbfindels. Beitr. path. Anat. 48, 499-525. FAHR, T. (1909). Ueber die Ganglien des menschlichen Herzens. Munch. med. Wschr. 56, 1509. FAHR, T. (1910). Zur Frage der Ganglienzellen im menschlichen Herzen. Zbl. Herz.- u. Gefdsskr. 6, 195-200. FIELD, E. J. (1951). The nervous component of the atrioventricular bundle. J. Anat., Lond., 85, 105-112. FUKUTAKE, K. (1924). Beitrage zur Histologie und Entwicklungsgeschichte des Herznervensystems. Z. ges. Anat. 1. Z. Anat. EntwGesch. 76, 592-639. GLOMSET, D. J. & BIRGE, R. F. (1945). A morphologic study of the cardiac conduction system. Part IV. The anatomy of the upper part of the ventricular septum in man. Amer. Heart ,J. 29, 526-538. GLOMSET, D. J. & BIRGE, R. F. (1948). A morphologic study of the cardiac conduction system. Part V. The pathogenesis of heart block and bundle branch block. Arch. Path. Lab. Med. 45,
135-170.

GLOMSET, D. J. & GLOMSET, A. T. A. (1940). A morphologic study of the cardiac conduction system in ungulates, dog and man. Part II. The Purkinje system. Amer. Heart J. 20, 677-701. HOLMES, A. H. (1921). The auriculoventricular bundle in mammals. J. Anat., Lond., 55, 269-285. HOLMES, W. (1947). The peripheral nerve biopsy. In Recent A4dvances in Clinical Pathology, pp. 404-405. Ed. S. C. Dyke. London: Churchill. HUBER, G. C. & DEWITT, L. Ml. H. (1898). A contribution on the motor nerve endings and on the nerve endings in the muscle spindles. J. Comp. Neurol. 7, 169-230. JACQUES, P. (1894). Recherches sur les nerfs du coeur chez la grenouille, et les mammiferes. J. Anat., Paris, 30, 622-648. JARISCH, A. & RICHTER, H. (1939). Die afferenten Bahnen des Veratrineffektes in den Herznerven. Arch. exp. Path. Pharrnak. 193, 355-371. JONES, T. (1927). Intramuscular nerve elements of the ventricular muscle. J. Anat., Lond., 61,
247-260.

KASEM-BECK, A. N. (1885). Zur Kenntnis der Herznerven. Arch. mikr. Anat. 24, 11-19. KASEM-BECK, A. N. (1887). Ueber das Vorkommen von Ganglien und einzelnen Nervenzellen auf den Herzventrikeln des Menschen, der Saugetiere und der Vogel. Zbl. med. Wiss. no. 42, 785-787.

Neurological studies of the cardiac ventricles of mammals

141

KING, A. B. (1939). Nerve endings in the cardiac muscle of the rat. Johns Hopk. Hosp. Bull. 65, 489-499. Kiss, F. (1932). Sympathetic elements in the cranial and spinal ganglia. J. Anat., Lond., 66, 488-498. LANDAU, E. (1950). Contribution a l'innervation du coeur (fibres myelinisees). Acta Anat. 9, 227-234. LAVRENTIEv, B. I. (1946). The innervation of the heart. Amer. Rev. Sov. Med. 3, 229-235. LAWRENTJEw, B. .J. (1927). Die Faserendigungen des N. vagus im Saugetierherzen. Anat. Anz. 64, 59-62. LAWRENT.JEW, B. .J. (1929). Experimentell-morphologische Studien fiber den feineren Bau des autonomen Nervensystems. I. Die Beteiligung des Vagus an der Herzinnervation. Z. mikr.anat. Forsch. 16, 383-411. LAWA-RENT.JEw, B. I. & GURWITSCH-LASOWSKAJA, A. S. (1930). Zur Frage der Innervation des Atrioventrikularbfindels His-Tawaras bei Saugetieren. Z. mikr.-anat. Forsch. 21, 585-596. LEE, H. (1849a). On the ganglia and nerves of the heart. Philos. Trans. part I, 43-46. Lili:, lH. (1849b). Postscript to a paper 'On the ganglia and nerves of the heart'. Philos. Trans. part I, 47-48. LE\IS. T. (1947). Pain. New York: Macmillan Co. 'MEIKLE.JOHN, T. (1913). On the innervation of the nodal tissue of the mammalian heart. J. Anal. Physiol. 48, 1-18. MEIKLEJOHN .J. (1914). On the topography of the intra-cardiac ganglia of the rat's heart. J. Anat. Physiol. 48, 378-390. MICIIAo1,01W, S. (1912). Innervation des Herzens im Lichte der netiesten Forschungen. Z. zeiss. Zool. 99, 539-687. Momisox, A. (1912). On the innervation of the sinoauricular node (Keith-Flack) and the auriculoventricular bundle (Kent-His). J. Anat. Physiol. 46, 319-327. MULLER, L. R. (1911). Beitrage zur Anatomie, Histologic und Physiologie des Nerven vagus, zugleich ein Beitrag zur Neurologie des Herzens, der Bronchien und des Magens. Dtsch. Arch. klin. Med. 111, 421-481. NETTLESIIIP, NV. A. (1936). Experimental studies on the afferent innervation of the cat's heart. J. Comp. Neurol. 64, 115-131. Noc, F. E. (1899). Atude anatomique des ganglions nerveux du coeur chez le chien et de leurs modifi cations dans l'intoxications diphtheritique exp~rimentale aigue. Bordeaux. (Cited by AMichailow, 1912.) NONIDEZ, .1. F. (1939). Studies on the innervation of the heart. I. Distribution of the cardiac nerves with special reference to the identification of the sympathetic and parasympathetic postgangylionics. Amer. .J. Anat. 65, 361-413. NONIDEZ, .J. F. (1943). The structure and innervation of the conductive system of the heart of the (og and Rhesuis monkey, as seen with a silver impregnation technique. limer. Heart J. 26,
577-597.

1PERMAN, E. (1924). Anatomische Untersuchungen fiber die Herznerven bei den hoheren Sdugetieren und beim M1enschen. Z. ges. Anat. 1. Z. Anat. EntwGesch. 71, 382-457.
I'LECHKOVA, E. K. (1936). On the morphology of the antagonistic innervation of the heart. Bull. Biol. Med. Exp. U.S.S.R. 1, 402-403. REMAK, 1I. (1838). Ueber die Berichtungen des organischen Nervensystems. Neue Notizen alS demn Gebiete der Natur- und Heilkunde (Froriep), 7, no. 137, 65-70. IREMAK, 11. (1839). Ueber die Ganglien der Herznerven des Menschen und deren physiologische Bedeutuing. Wschr. ges. Heilk. no. 10, 149-154. IREMAK, 11. (1844). Neurologische Erlaiuterungen. Muller's Arch. Anat. Physiol. pp. 463-472. RETZER, It. (1908). Some results of recent investigations of the mammalian heart. Anal. Rec. 2, 149-155. RIchITER, H. & AMANN., A. (1940). Ist die Lunge chemosensibel? Arch. exp. Path. Pharmak. 196,
274-279.

SCAGLIA, G. (1927). L'apparato nervoso contenuto nel sistema atrioventricolare di Bos tauris. Arch. ital. Anat. Embriol. 24, 658-696.
SCARPA, A. (1794). Tabulae Neurologicae. Ticini. SCHKLAREWSKI, A. (1872). Ueber die Anordnung der Herzganglien bei Vogeln und Saugethieren. Nachr. Ges. Wiss. Gottingen, pp. 426-428. (Cited by Michailow, 1912.)

142

Francis Davies, E. T. B. Francis and T. S. King

SIMPsON, G. G. (1945). The principles of classification and a classification of mammals. Bull. Amer. Mus. nat. Hist. 85, 1-350. SINGER, R. (1926). Experimentelle Studien uber die Schmerzempfindlichkeit des Herzens und der grossen Gefaisse und ihre Beziehung zur Angina pectoris. Wien. Arch. inn. Med. 12,
193-212. SKWORZOW. (1874). Zur Frage uber die Anatomie und Histologie des Herzens und Herzbeutels. Dissert. Petersburg. (In Russian. Cited by Michailow, 1912.) SMIRNOW, A. (1895). Ueber die sensibilen Nervenendigungen im Herzen bei Amphibien und Saugetieren. Anat. Anz. 10, 737-749. SMIRNOW, A. (1905). Einige Bemerkungen uber die Existenz von Ganglienzellen in den Herzventrikeln des Menschen und einiger Saugetiere. Arb. anat. Inst., Wiesbaden, 27, Abt. 1, Hft. 81, 295-301. SODI-PALLARES, D., RODRIGUES, M. I., CHAIT, L. 0. & ZUCKERMANN, R. (1951). The activation of the interventricular septum. Amer. Heart. J. 41, 569-608. STOHR, P. JR. (1928). Die peripherischen Anteile des vegetativen Nervensystems. In M6llendorff's Handbuch der mikroskopischen Anatomie des Menschen, Bd. 4, T. 1, 265-447. Berlin: Springer. ST6HR, P. JR. (1932). Nerves of the blood vessels, heart, meninges, digestive tract and urinary bladder. In Penfield's Cytology and Cellular Pathology of the Nervous System, vol. 1, sect. 8, 381-420. New York: Hoeber. STOTLER, W. A. & McMAHON, R. A. (1947). The innervation and structure of the conductive system of the human heart. J. Comp. Neurol. 87, 57-83. SUTTON, D. C. & LUETH, H. C. (1930). Pain. Arch. intern. Med. 45, 827-867. TAWARA, S. (1906). Das Reizleitungssystem des Saugetierherzens. Jena: Fischer. TCHENG, K. T. (1951). Innervation of the dog's heart. Amer. Heart J. 41, 512-524. TRUEX, R. C. & COPENHAVER, W. M. (1947). Histology of the moderator band in man and other mammals with special reference to the conduction system. Amer. J. Anat. 80, 173-201. VALEDINSKY, I. A. (1905). Zur Frage uber die Nervenknoten im Herzventrikel einiger Saugetiere. Arb. anat. Inst., Wiesbaden, 27, 285-294. VALEDINSKY, A. (1908). Zur Frage fiber die Nervenknoten im Herzventrikel einiger Saugetiere. Inaug. Diss., Tomsk. (In Russian. Cited by Michailow, 1912.) VALEDINSKY, A. (1910). Einige Erganzungen zur Frage nach der Gegenwart und der Verteilung der Nervenganglien in den Herzkammern einiger Saugetiere und des Menschen. Anat. Anz.

37, 465472. VIGNAL, W. (1881). Recherches sur l'appareil ganglionnaire du coeur des vertkbres. Cole prat. d. hautes etudes. Lab. d'histol. du Coll. de France. Trav., Paris, 6, 186-254. VITALI, G. (1937). Contributo allo studio del sistema muscolare specifico del cuore. Le espansiono nervose nelle fibre di Purkinje in Ovis aries. Anat. Anz. 84, 88-102. WAHLIN, B. (1935). Das Reizleitungssystem und die Nerven des Saugetierherzens. Inaug. Diss. Isaac Marcus Boktryckeri-Aktieboalg, Stockholm. WEDDELL, G. &. ZANDER, E. (1950). A critical evaluation of methods used to demonstrate tissue neural elements, illustrated by reference to the cornea. J. Anat., Lond., 84, 168-195. WEINREIcH, M. (1888). Ueber Nerven und Ganglienzellen im Saugetierherzen. Inaug. Diss., Halle. (Cited by Michailow, 1912.) WILSON, J. G. (1909). The nerves of the atrioventricular bundle. Proc. roy. Soc. B, 81, 151-164. WOLHYNSKI, T. (1928). Innervation des Herzkammer- und Vorhofsseptum des Kalbes. Z. ges. Anat. 1. Z. Anat. EntwGesch. 86, 608-638. WOOLLARD, H. H. (1926). The innervation of the heart. J. Anat., Lond., 60, 345-373.
EXPLANATION OF PLATES PLATE 1 Fig. 1. Ventral surface of heart of calf showing disposition of epicardial nerve cells. x I natural size. Fig. 2. Dorsal surface of heart of calf showing disposition of epicardial nerve cells. x I natural size. Fig. 3. Interior of right atrium and ventricle of heart of calf. The dots indicating the position of subendocardial nerve cells obviously follow the path of the A.v. node, the A.V. bundle and the proximal part of its right limb. x I natural size.

Journal of Anatomy, Vol. 86, Part 2

Plate

......

,.
*36

... as.

.01
:0

0..

I.e.

r
.4F p*

DAVIES, FRANCIS

AND

KING-NEUROLOGICAL STUDIES

OF THE CARDIAC VENTRICLES OF MAMMALS

Plate 2

Journal of Anatomy, Vol. 86, Part 2

t..

, 4!1$

AO

*-

.N

or*...

t' '

V'i
is

^;

DAVIES, FRANCIS AND KING-NEUROLOGICAL

STUDIES OF THE CAR1)IAC VENTRICLES OF MAMMALS

Journal of Anatomy, Vol. 86, Part 2

Plate 3

,3 't..

:. By.

......

.,!!.4

17

DAVIES, FRANCIS

ANi)

KING

NlwUlROLOGICAL STUDIES

OF

TIlE

CARDIAC VENTRICLES

01F MAMMALS

Plate 4

qi *,.*'< ait!
S# ' ' '@e ~~~~~~~~~~~~~~~~~~~~~~~~~~S

@wfri

;YS ~

':. . ~~~~~~~~~~~~~~~~~~~4
4S24.

.i_X~ ~w.
* .* 4

Journal of Anatorny, Vol. 86, Part 2

iXA4*M'.

t<

e>s
s'vJ ~~~~~~~~~~~~i r B~:1

l~

<

DAVIES, FRANCIS

AND KING

NEUROLOGICAL STUDIES OF THE CARDIAC VENTRICLES OF MAM1MALS

Neurological studies of the cardiac ventricles of mammals

143

Fig. 4. Interior of left atrium and ventricle of heart of calf. The dots indicating the position of subendocardial nerve cells obviously follow the path of the left limb of the A.V. bundle and some of its major branches. x i natural size.
PLATE 2 Fig. 5. Nerve cells in course of nerve trunk in epicardium of right ventricle of calf. Silver-pyridine.
x 100.

Fig. 6. Nerve cells forming ganglion alongside nerve trunk in epicardium of left ventricle of calf. Silver-pyridine. x 62. Fig. 7. Medullated and non-medullated nerve fibres in epicardium of right ventricle of cow. Osmic acid. x 260. Fig. 8. Bipolar nerve cell in epicardium of left ventricle of cow. Bielschowsky-Gros. x 350. Fig. 9. Nerve cells alongside left limb of A.v. bundle of calf. Silver-pyridine. x 112. Fig. 10. Unipolar nerve cells in epicardium of right ventricle of calf. Bielschowsky-Gros. x 400. Fig. 11. Multipolar nerve cells in A.v. bundle of cow. Silver-pyridine. x 480. Fig. 12. Unipolar cells in epicardium of right ventricle of cow. Note that the central nerve cell is surrounded by satellite cells and that its process passing to the right divides into two a short distance from the cell body. Bielschowsky-Gros. x 435.
PLATE 3 Fig. 13. Nerve cells in epicardium of left ventricle of cow. Note nerve fibre (derived from another cell) winding around the left and right processes of the multipolar cell in the lower part of the field. Bielschowsky-Gros. x 440. Fig. 14. Nerve endings in the form of small knobs in relation to nerve cells and their processes. Epicardium of left ventricle of cow. Holmes silver. x 410. Fig. 15. Nerve cells in epicardium of left ventricle of calf. Note two processes arising conjointly from upper surface of the lowermost cell. Bielschowsky-Gros. x 390. Fig. 16. Simple nerve endings in epicardium of right ventricle of calf. Holmes silver. x 900. Fig. 17. Nerve fibre forming spiral around individual muscle fibre in left ventricle of calf. Bielschowsky-Gros. x 690. Fig. 18. Plexus of nerve fibres on surface of Purkinje fibres in A.V. bundle of cow. Silver-pyridine. x 420. Fig. 19. Simple nerve endings on surface of Purkinje fibres in A.V. bundle of cow. Silver-pyridine. x 410. PLATE 4 20. A.V. bundle of dog. Note relative paucity of nerves. Holmes Silver. x 164. Fig. Fig. 21. Part of previous figure under higher magnification. x 300. Fig. 22. Upper part of right limb of A.v. bundle of dog to show complete absence of nerve fibres. Note differentiation of muscle fibres of the right limb from those of the surrounding myocardium. Holmes silver. x 340. Fig. 23. A.v. bundle of cow. Note profusion of nerves. Holmes silver. x 92. Fig. 24. Note nerve ending in right part of figure among muscle cells of the media of a small coronary artery. Dog. Holmes silver. x 1220. Fig. 25. Note nerve endings in adventitia of a coronary artery. Dog. Holmes silver. x 920. Fig. 26. Note nerves ending among muscle cells of the media of a coronary artery. Dog. Holmes silver. x 620.